13 Replies Latest reply on Nov 25, 2010 2:44 PM by Geoffrey Hynes

    Adding Datasets to V2

    New User

      I notice that the format of the datasets in V2 is slightly different from that in V1. When importing a dataset in V2 the wizard asks for details of endpoints, rather than the user being able to specify a SIDS path within the data. Does this mean that only single endpoint datasets can be imported? I have successfully imported datasets by splitting my multi endpoint sets into individual single endpoint sets; is this the right approach?

      Thanks,

      Nick

        • Re: Adding Datasets to V2
          New User

          Hi Nick,

           

          The new version still has the ability to import databases using either the vertical or horizontal method. The vertical method does not give you the additional fields for say entering comments, which I find is a very useful field. Especially when I want to clarify the results obtained by a third party or it's a very old study. The horizontal method does still allow the manual highlighting of the SIDS path. From my initial playing around with the importing of databases and the brief demonstration from Ovanes Mekenyan a couple of weeks ago, your company specific database should import smoothly without additional changes and you should be able to highlight the SIDS position.

          All the best,

          Geoff...

            • Re: Adding Datasets to V2
              New User

              Geoff,

              Of course, silly of me; I've done it lots of times in v1 but failed to notice the "Vertical" and "Horizontal" buttons in V2!

              However, that said I'm still having problems in that when I try to enter my new dataset using exaclty the same Excel sheet as with V1 but with the V2 Endpoint paths copied (where different from V1). I have attached a snip of the Import Wizard window. In this particular instance I'm importing Daphnia magna EC50 data which is a continuous value so I selected Mean value/Scale value. Although I get the "Import successful" message, the error window pops up with the message "Invalid endpoint tree position" for every row and no data is imported.

                • Re: Adding Datasets to V2
                  83059 Expert

                  The structure of the data tree has been changed compared to version 1, due to the fact that the data tree is now configurable. So it is likely that you have to redefine the endpoint path and change it in your excel file.

                   

                  The revision of the guidance document on how to import your own dataset is very high on our priority list.

                    • Re: Adding Datasets to V2
                      New User

                      Hi Bob,

                       

                      Just a couple of issues here. May be something I'm doing and I've set aside a couple of hours to really nail this tomorrow.

                      However, comments so far:

                       

                      Having now used a propriety database and not a supplied dataset, the “horizontal” import does not seem to be anywhere near as user friendly as before.

                       

                      In fact, having spent a large amount of time generating a large database of propriety structures, I’m more than a bit annoyed to see changing such as “Toxicological Information” now changed to “Human health hazards”. Why the change? Not much consideration for those of us who have generated large databases.

                       

                      The SIDS path before was:

                      Toxicological Information# Genetic Toxicity (mutation and chromosomal aberrations)# In Vitro# Chromosomal aberration# Chromosomal aberration with S9#Chinese Hamster Ovarian Cells

                       

                      Now we have:

                      Human health hazards#Genetic Toxicity#in vitro#in vitro mammalian chromosome aberration test#Chromosome aberration#Undefined Test organisms (species)#without S9

                       

                      “Environmental Fate” has also changed to “Environmental Fate and Transport”. Why?

                       

                      On the examples I’ve chosen, my old import file (Excel) went from “A to X”. On exporting a dataset for cyclophosphamide, the export file goes from “A to DK”. An example of too much information or over complication? Or data not actually required. Do we need such a large range of information, or can we just configure our database to the actual fields we would like to use?

                       

                      Interestingly, if you use the horizontal import and forget to select the data region, an error message is given after you hit the finish button, then the Toolbox closes down. Checked and this occurred 1 out of 3 times.

                      (error screen dump attached)

                       

                      Still haven’t had enough time to successfully import my database. Will keep trying.

                       

                        • Re: Adding Datasets to V2
                          New User

                          Geoff,

                          From looking at an exported dataset it appears that the endpoint position (previously called SIDS) must be spread across columns instead of simply pasting the copied endpoint position into a single column as previously. So the new path would look like this;

                          Ecotoxicological InformationAquatic ToxicityMortalityLC5023 DaysAnimaliaArthropoda(Invertebrates)Branchiopoda(branchiopods)Daphnia magna

                           

                          as opposed to the old version which had hashes between the fields but could be pasted into a single spreadsheet column.

                          Many of the other fields are, I believe optional

                          Haven't tried this yet so hoping it works.

                           

                          Nick

                            • Re: Adding Datasets to V2
                              New User

                              Hi Nick,

                               

                              This is the format of my databse which is already separated and works for TB v1.

                              I originally separated the endpoints as I found this was easy to work with and gave better clarity to each endpoint, especially when some endpoint path names are very similar.

                               

                              SIDS position
                              Toxicological InformationGenetic Toxicity (mutation and chromosomal aberrations)In VitroAMES_MutagenicityAMES Mutagenicity with S9ratAMES-Salmonella Typhimurium TA 98
                              Toxicological InformationGenetic Toxicity (mutation and chromosomal aberrations)In VitroAMES_MutagenicityAMES Mutagenicity with S9ratAMES-Salmonella Typhimurium TA 100
                              Toxicological InformationGenetic Toxicity (mutation and chromosomal aberrations)In VitroAMES_MutagenicityAMES Mutagenicity with S9ratAMES-Salmonella Typhimurium TA 1535
                              Toxicological InformationGenetic Toxicity (mutation and chromosomal aberrations)In VitroAMES_MutagenicityAMES Mutagenicity with S9ratAMES-Salmonella Typhimurium TA 1537
                              Toxicological InformationGenetic Toxicity (mutation and chromosomal aberrations)In VitroAMES_MutagenicityAMES Mutagenicity with S9ratAMES-Salmonella Typhimurium TA 1538

                               

                              When importing I have highlighted the relevant fields which I wanted the TB to pick up and import, however this still didn't work.

                               

                              The fields chosen are included in the attached MS Word doc.

                                • Re: Adding Datasets to V2
                                  New User

                                  Hi,

                                   

                                  I’ve started from the beginning again and found a few very annoying examples already.

                                  I still could not import melting point. These are not small simple change to importing of databases.

                                   

                                  For inclusion in to the TB, an exact match must be made (lower case and capitals). Why then have the following been changed ? and these are the easy ones to see as they have short path names.

                                   

                                  Why have the following been changed ?

                                   

                                  Physical Chemical Properties#Melting point

                                  to

                                  Physical Chemical Properties#Melting / freezing point

                                   

                                  And

                                   

                                  Physical Chemical Properties#Boiling Point

                                  to

                                  Physical Chemical Properties#Boiling point

                                   

                                  And

                                   

                                  Physical Chemical Properties#Water Solubility

                                  to

                                  Physical Chemical Properties#Water solubility

                                   

                                  Where is the consistency we expect to help aid importation as there will be quite a few new users trying to import large datasets. Exactness is the difference between successfully importing propriety data, or not.

                                   

                                  It would be extremely helpful to be able to export the complete table of every eventuality (Excel file), as at the moment a data point value is required in the relevant field to be able to then export the path string.

                                  You could do this with v1 of the TB which was very helpful.

                                    • Re: Adding Datasets to V2
                                      83059 Expert

                                      Hi Geoffrey,

                                       

                                      we're still looking into the other issues, but I can already give you the background behind the label changes.

                                       

                                      This is due to harmonisation with the OECD Harmonised Templates [see www.oecd.org/ehs/templates].

                                       

                                      Indeed these Harmonised templates are the OECD recommended database structure for summary information for study results. It was therefore logical to implement them in the Toolbox. It's not only the labels that were changed. The whole database structure was changed to comply with the OECD Harmonised Templates. This is what now allows the import/export with IUCLID. It will also allow in the future data exchange with eChemPortal.

                                       

                                      We'll look into the other issues and see if we can provide some guidance.

                                       

                                      hang in there

                                       

                                      Bob

                                        • Re: Adding Datasets to V2
                                          New User

                                          Hi Bob,

                                           

                                          Thanks for the clarification.

                                           

                                          However, I've now simplified by database by removing both columns and all data apart from PhyChemProp and also even tried to import as an inventory. I still receive an error message.

                                          I have included my import Excel file below, it may be something I'm just not doing correctly.

                                          Having said that, I didn't have any issues with v1 of the TB.

                                           

                                          Importing:

                                          The only 3 data points not automatically picked up the the importing system are:

                                          Chemical name

                                          Data

                                          Endpoint Tree Path

                                           

                                          All other data points are automatically selected.

                                            • Re: Adding Datasets to V2
                                              83059 Expert

                                              I've asked LMC to look into it.

                                               

                                              They promised to prepare a guidance document by next Wednesday.

                                                • Re: Adding Datasets to V2
                                                  New User

                                                  Hi Bob,

                                                   

                                                  Just some additional information.

                                                   

                                                  I have successfully exported a know chemical mutagen used in the rat bone micronucleus test called cyclophosphamide (CAS 000050-18-0). I then changed the CAS number, chemical name and Smiles field to that used previously. This data should be in a form that is acceptable for importation, however, I recieved the same "invalid value" error for all 229 data points.

                                                   

                                                  Import error_02.jpg

                                                  I hope this is useful and helps,

                                                  Geoff...

                              • Re: Adding Datasets to V2
                                83059 Expert

                                Dear all,

                                 

                                as promised, please find enclosed a drft guidance on importing databases into version 2.0.

                                 

                                All comments are welcome.

                                 

                                best regards

                                 

                                Bob